Replication-based outsourced computation enables efficient correctness verification through cross-checking results from multiple non-colluding clouds, yet introduces critical privacy risks when clouds exchange intermediate data for interest alignment, especially in multi-user private data scenarios such as federated learning. To address this challenge, we propose BDT-RVOC, a novel framework integrating binary-segmentation data truncation with distributed multi-trapdoor public key cryptography (DMT-PKC). BDT-RVOC dynamically splits raw data into mutually exclusive blocks distributed to different clouds, ensuring that computations on truncated data preserve bit-level consistency with raw-data operations while preventing raw-data leakage during inter-cloud exchanges. Our framework designs secure interactive protocols for addition, multiplication, and equivalence testing under truncation, formally guaranteeing operational equivalence between block-level and raw-data computations. By extending Paillier encryption to a multi-key setting, DMT-PKC partitions strong private keys to resist collusion between the aggregator and up to n−2 agents. Applied to private aggregation, BDT-RVOC achieves a 1.8× speedup over federated learning baselines with 57% lower communication overhead. Theoretical analysis proves resilience against semi-honest adversaries compromising all communications and up to n−1 colluding parties, while experiments on real-world datasets confirm practical efficiency and scalability for real-world deployments.

In the dynamic moving process of vehicle clusters, there are several challenges, including inefficiencies, cross-domain trust issues and privacy leakage. To address these issues, we propose a blockchain-assisted group and cross-domain authentication key agreement, which implements distributed key management based on a threshold key sharing scheme, and realizes group authentication and group key distribution for vehicle clusters through the collaboration of roadside units (RSUs) and Key Generation Center (KGC). Meanwhile, a cross-domain trust chain is constructed based on blockchain to accomplish secure cross-domain authentication and key agreement without the participation of the original KGC, which solves the problem of trust deficiency and single-point vulnerability in the process of cross-domain communication. Finally, we employed Real-or-Random (ROR) formal security analysis and the ProVerif tool to verify that the proposed authentication scheme, the results show that the proposed scheme is secure and superior to existing schemes in terms of communication and computational overhead.

The Internet of Vehicle (IoV) is a core component of smart city development. However, data interactions between IoV entities involve personal privacy, and once maliciously attacked, they may threaten the stable operation of the entire transportation system. Traditional authentication protocols suffer from high computational and communication overheads and are vulnerable to various threats, including physical attacks, entity impersonation, and replay attacks. Moreover, due to their reliance on centralized trusted authority(TA), traditional protocols are prone to single-point failures, especially when handling large-scale vehicle access. To address these challenges, this paper proposes a lightweight mutual authentication protocol based on physical unclonable function(PUF), which not only ensures vehicle anonymity and traceability but also supports a pseudonym update function after authentication. The protocol employs an architecture in which the main TA(MTA) is responsible for registration and data storage, while the sub-TA(STA) handles authentication, thereby effectively mitigating the risk of a single-point failure. Additionally, to counter the exposure of a large number of challenge-response pairs in traditional PUF-based authentication—making them susceptible to machine learning(ML)-based modeling attacks—this paper introduces a CUBE-PUF scheme based on digital Rubik’s Cube and random numbers. This approach enhances response unpredictability and randomness. We conduct both formal and informal security analyses of the proposed protocol and rigorously verify its security using the ProVerif verification tool. Furthermore, comparative evaluations with existing protocols demonstrate that our approach significantly reduces communication and computational overhead while offering enhanced security.

Delegation Learning(DL) flourishes data sharing, enabling agents to delegate data to the cloud for model training. To preserve privacy, homomorphic encryption (HE) offers an effective solution for privacy-preserving machine learning (PPML) in delegation learning, yet faces critical challenges in functionality (non-linear activation support), practicality (ciphertext blow-up from iterative computations), and security (data leakage risks caused by public knowledge of the mathematical principles applied in model training). To tackle these challenges, we propose THC-DL, a three-party HE framework addressing these challenges holistically for the first time. We elaborately design the ciphertext secure comparison (DL-CSC) protocol to satisfy secure comparison with private inputs, enabling efficient non-linear operations with O(1) communication complexity that reduces runtime to 12.5% of the DGK (Dolev-Greensmith-Kent protocol, the well-known comparison protocol). Second, we construct a Truncation-Mapping (Tru-Map) scheme, a mechanism that transforms input data by truncating and mapping it into a domain that facilitates more efficient processing, and the addition of truncated mapped data to resolve ciphertext blow-up by adaptively scaling ciphertexts during iterative training, ensuring correctness. Third, we formalize data leakage risks in HE-based quadratic convex optimization (standard in ML) and apply THC-DL to construct a secure optimization scheme. Theoretical analysis confirms THC-DL’s resilience against input recovery attacks, even when adversaries exploit public model parameters. Experiments on a real-world platform validate DL-CSC’s efficiency and scalability while reducing the computational complexity and communication complexity from O(n) to O(1) where n denotes the length of the input bits.

In this paper, we leverage unmanned aerial vehicles (UAVs) to enhance mobile crowd sensing (MCS) by addressing two critical challenges: uncontrollable data quality and inevitable unsensed points of interest (PoIs). We introduce a UAV-assisted method to deal with these challenges. To ensure the accuracy of sensing data contributed by human participants, the proposed truth discovery method utilizes UAV-collected sensing data as few-shot samples to train the truth discovery model, which is then employed to calibrate sensing data solely collected by human participants. Additionally, to meet the sensing coverage requirement, we present a method that predicts data values for unsensed PoIs by utilizing their historical sensing data and the sensed neighboring PoIs information. The method employs a graph neural network to capture spatio-temporal relationships of the sensing data, facilitating accurate estimation of unsensed PoIs. Through extensive simulations, our approaches demonstrate superior performance compared to existing methods, showcasing the potential of UAV-assisted MCS for overcoming challenges and enhancing data collection efficiency in various domains.

Recent research has increasingly focused on multiple intent detection and slot filling tasks due to the closer to the complex multi-intent scenarios in the real world. However, existing approaches face two potential issues: (i) Single joint models are unable to capture full information and handle complex relations in data, and (ii) Irrelevant label relations can lead to over-guidance when conducting explicit interactions in multi-intent scenarios. To address the above issues, we propose a hybrid model based on Graph Convolutional Network and Multi-head Transformer (GCN-MT). This model alleviates over-guidance through a graph network at the entire corpus level and realizes intent-slot interaction through a co-attention network at both the specific utterance and local token levels. Experimental results demonstrate that our approach outperforms existing reproducible models on two public multi-intent datasets.

The Internet of Vehicles (IoV) is a specific instance of the Internet of Things (IoT) in the transportation field, driven by application requirements, such as intelligent traffic services and automatic vehicle control, can improve road safety and enhancing transmission efficiency. However, highly open networks tend to bring more security threats, and secure authentication becomes an important guarantee for reliable communication. Traditional IoT authentication and key agreement methods are costly, inefficient, and rely on the third-party trusted institutions, making them unsuitable for direct application in IoV systems. To meet the security authentication needs of IoV, and improve authentication efficiency and anonymity, this article proposes a verifiable commitment-based mutual authentication and key agreement protocol, called mutual authentication and key agreement protocol based on verifiable commitment (VC-MAKA). In VC-MAKA, we construct a verifiable commitment scheme where the verifier can verify the committed secret. Furthermore, based on this verifiable commitment scheme, we implement secure authentication and session key agreement, allowing vehicles to freely negotiate secure session keys and achieving conditional anonymous protection. Additionally, the proposed VC-MAKA also achieves rapid session key updates, enhancing the security of the session keys. We have conducted formal and informal security analysis, and the results show that VC-MAKA meets security requirements, such as mutual authentication, anonymity, traceability, and untraceability. Moreover, we have used the ProVerif tool for security experiment and performance comparison analysis, and the results indicate that compared to other schemes, the VC-MAKA protocol offers higher security and better efficiency.

With the outbreak of COVID-19, the government has been forced to collect a large amount of detailed information about patients in order to effectively curb the epidemic of the disease, including private data of patients. Searchable encryption is an essential technology for ciphertext retrieval in cloud computing environments, and many searchable encryption schemes are based on attributes to control user's search permissions to protect their data privacy. The existing attribute-based searchable encryption (ABSE) scheme can only implement the situation where the search permission of one person meets the search policy and does not support users to obtain the search permission through collaboration. In this paper, we proposed a new attribute-based collaborative searchable encryption scheme in multi-user setting (ABCSE-MU), which takes the access tree as the access policy and introduces the translation nodes to implement collaborative search. The cooperation can only be reached on the translation node and the flexibility of search permission is achieved on the premise of data security. ABCSE-MU scheme solves the problem that a single user has insufficient search permissions but still needs to search, making the user's access policy more flexible. We use random blinding to ensure the confidentiality and security of the secret key, further prove that our scheme is secure under the Decisional Bilinear Diffie-Hellman (DBDH) assumption. Security analysis further shows that the scheme can ensure the confidentiality of data under chosen-keyword attacks and resist collusion attacks.

The recently proposed DEtection TRansformer (DETR) achieves promising performance for end-to-end object detection. However, it has relatively lower detection performance on small objects and suffers from slow convergence. This paper observed that DETR performs surprisingly well even on small objects when measuring Average Precision (AP) at decreased Intersection-over-Union (IoU) thresholds. Motivated by this observation, we propose a simple way to improve DETR by refining the coarse features and predicted locations. Specifically, we propose a novel Coarse-to-Fine (CF) decoder layer constituted of a coarse layer and a carefully designed fine layer. Within each CF decoder layer, the extracted local information (region of interest feature) is introduced into the flow of global context information from the coarse layer to refine and enrich the object query features via the fine layer. In the fine layer, the multi-scale information can be fully explored and exploited via the Adaptive Scale Fusion(ASF) module and Local Cross-Attention (LCA) module. The multi-scale information can also be enhanced by another proposed Transformer Enhanced FPN (TEF) module to further improve the performance. With our proposed framework (named CF-DETR), the localization accuracy of objects (especially for small objects) can be largely improved. As a byproduct, the slow convergence issue of DETR can also be addressed. The effectiveness of CF-DETR is validated via extensive experiments on the coco benchmark. CF-DETR achieves state-of-the-art performance among end-to-end detectors, e.g., achieving 47.8 AP using ResNet-50 with 36 epochs in the standard 3x training schedule.

Time series forecasting has historically been a popular research area, attracting widespread interest in academia and industry. Recently, Deep Learning based models like RNN, LSTM, and NeuralProphet have been successfully applied for time series forecasting. However, single forecasting models are unable to capture full information and learn complex patterns in data. The combination of models has proved to be an effective strategy to address this issue. Traditional hybrid structures still limit the forecasting ability of hybrid methods. In this paper, we propose a novel hybrid forecasting model based on LSTM and NeuralProphet (NP-LSTM), which is constructed by a parallel-series hybrid structure. The proposed model uses LSTM to model diverse nonlinear relationships and NeuralProphet is designed to extract primary trends and seasonal effects and provide interpretability. In the experiments of this paper, we compare our model with other single models and hybrid models of traditional structures using four real-world datasets. The experimental results show that our NP-LSTM hybrid model obtains superior performance in various metrics for time series forecasting.

Automatic speaker verification (ASV) is to verify the identity of speaker from a given speech utterance without direct supervision from outside entities. Majority of recent ASV systems with deep speaker embedding apply temporal pooling or similar techniques for frame-level feature aggregation in time domain. In this paper, we propose a deep speaker embedding network for adaptively modelling and fusing multi-part information in frequency-time domain, using a modified ResNet-SO to encode acoustic features into global information, a proposed multi-part information aggregator to distinguish global information and different part features for aggregating them with adaptive weight pooling to unified utterance-level embedding descriptors. More-over, we design a privacy-preserving manner and preliminarily implement it in prototype system. Experiments are conducted on three scale datasets. We demonstrate that the presented multi-part information aggregator with adaptive weight pooling is superior for producing discriminative and robust utterance-level embedding descriptors. We also show that our network achieves state-of-the-art performance by a significant margin on the popular VoxCelebl while requiring fewer parameters than previous approaches.

Text-independent speaker verification (TI-SV) refers to the process of verifying an individual’s claimed identity from a given speech utterance with unfixed content. Most deep speaker embedding networks of TI-SV apply temporal pooling or similar techniques for frame-level feature aggregation, and adopt a single loss function for training. In this paper, we propose a powerful hybrid network, named HN-MFML, which consists of a backbone and two sub-networks: speaker embedding extraction and speaker classification. The hybrid network not only incorporates global and local features, but also assigns adaptive weights to them. It adopts a modified ResNet-50 as backbone, using speaker embedding extraction sub-network to aggregate global and local features adaptively in frequency-time domain, which can be trained end-to-end by a loss. In addition, we add a speaker classification sub-network with another loss and explore a multi-loss fusion to jointly train for improving generalization. We demonstrate that our multi-feature aggregation and multi-loss fusion are superior for obtaining discriminative utterance-level embedding descriptors. We also show that HN-MFML achieves state-of-the-art performance by a significant margin compared with previous methods.

Tong Zhou, Pengfei Cao, Yubo Chen, Kang Liu, Jun Zhao, Kun Niu, Weifeng Chong, Shengping Liu. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Benefiting from the development of smart urban computing, the mobile crowd sensing (MCS) network has emerged as momentous communication technology to sense and collect data. The users upload data for specific sensing tasks, and the server completes the aggregation analysis and submits to the sensing platform. However, users’ privacy may be disclosed, and aggregate results may be unreliable. Those are challenges in the trust computation and privacy protection, especially for sensitive data aggregation with spatial information. To address these problems, a verifiable location-encrypted spatial aggregation computing (LeSAC) scheme is proposed for MCS privacy protection. In order to solve the spatial domain distributed user ciphertext computing, firstly, we propose an enhanced-distance-based interpolation calculation scheme, which participates in delegate evaluator based on Paillier homomorphic encryption. Then, we use aggregation signature of the sensing data to ensure the integrity and security of the data. In addition, security analysis indicates that the LeSAC can achieve the IND-CPA indistinguishability semantic security. The efficiency analysis and simulation results demonstrate the communication and computation overhead of the LeSAC. Meanwhile, we use the real environment sensing data sets to verify availability of proposed scheme, and the loss of accuracy (global RMSE) is only less than 5%, which can meet the application requirements.

For smartphone service providers, it is of vital importance to recognize characteristics of customers. The process of recognizing these characteristics is generally referred to as user profile, which provides knowledge basis for business decisions, enables intelligent services, and brings unique competitiveness. As a basic component of user profile, bedtime could reflect lifestyle, health condition, and occupation of people. This paper presents a flexible algorithm named BTP (Bedtime Prediction), which is designed for predicting wake time and bedtime by analysing screen status of smartphone. BTP first collects screen status log data of user’s smartphone and conducts preprocessing with a series of auxiliary user profiles. Then, it detects and records users’ wake time and bedtime of one day by searching and combining major screen extinguish periods in the past 24 hours. Finally, BTP predicts future bedtime by matching current screen status sequence with all historical records. By applying BTP, most of night and morning scenario‐based applications could provide more considerate services, rather than following fixed execution time like alarm clock. Experiments on practical applications prove that BTP can effectively predict wake time and bedtime without applying complicated machine learning algorithms or uploading data to server.

Opportunistic offloading can be utilized to offload computing tasks and traffic data in Mobile Edge Computing (MEC). To improve the ratio of successful data offloading and reduce unnecessary data redundancy in opportunistic forwarding process, some methods of evaluating a device’s forwarding capability are proposed. However, most of these methods do not consider the temporal impact from device mobility and the efficiency influence from the capability computation process. To settle these problems, we proposed a Transient‐cluster‐based Capability Evaluation Method (TCEM) to evaluate a device’s data forwarding capability. The TCEM can be divided into two steps. The first step aims to reduce computational complexity by evaluating a device’s possibility of contacting the destination within a time constraint based on the transient cluster generated by our proposed Transient Cluster Detection Method (TCDM). The second step is to calculate a device’s probability of directly and indirectly forwarding data to the destination. The probability as a metric of evaluating a device’s forwarding capability can be used in different data forwarding strategies. Simulation results demonstrate that the TCEM‐based data forwarding strategy outperforms other data forwarding strategies from the aspect of the proportion of the data delivery ratio to the data redundancy.